Adapting depression collaborative care models to increase uptake of computerized cognitive behavioral therapy at the...

To examine the feasibility, acceptability, and potential health effects of computerized cognitive behavioral therapy-enhanced collaborative care (cCBT-CC) versus usual primary care (UC).

Internet-based cCBT can effectively treat depression but is not widely used, including in the Veterans Health Administration where it was freely available for veterans. We adapted pre-existing depression collaborative care models using implementation and user-centered design strategies to facilitate cCBT implementation.

This pilot randomized controlled trial (RCT) included 57 VA primary care patients to cCBT-CC or UC. Participants had Patient Health Questionnaire (PHQ-9) scores of 10+. Those with serious mental illness (e.g., bipolar depression, schizophrenia) and active suicidality were excluded. Intervention patients received tailored Vets Prevail cCBT accompanied by collaborative care manager support, overseen by psychiatry and primary care. UC offered collaborative care services and digital mental health tools at baseline. Feasibility (patient reach, provider adoption, intervention implementation), acceptability (CSQ-8), and potential effectiveness (PHQ-9) data was collected at baseline and 3-months by a blinded study team member.

Participants (cCBT-CC n = 29, UC n = 28) were 50 years old (mean); 70 % men; 32 % White, 32 % Hispanic, 25 % Black; 21 % homeless-experienced. Mean baseline PHQ-9 scores were 15.1 (SD = 5.0); 39 % reported suicidal thoughts/behaviors. 72 % of 94 primary care providers, from 6 out of the 8 participating clinics, helped support their patients' participation. cCBT-CC participants received 4 care manager check-ins over 33 days totaling 113 min (64 % clinical; 36 % technical), on average. They completed mean 6.7 out of 11 cCBT lessons. Participants in the cCBT-CC arm experienced a statistically (not clinically) significant decline in the primary outcome of depression (Δ = -2.5; p = 0.02) symptoms from pretreatment to posttreatment. There was a greater, albeit non-significant, decrease in PHQ-9 scores among cCBT-CC participants over 3-months, compared to UC participants (Δ = -2.8; 95 % CI = -5.6, -0.01; p = 0.05).

cCBT-enhanced collaborative care appeared feasible, acceptable, and possibly effective in treating primary care patients with depression.